American Mineralogist, Volume 92, pages 1303–1315, 2007

Laser Raman microspectrometry of metamorphic quartz: A simple method for comparison

of metamorphic pressures
MASAKI ENAMI,1,* TADAO NISHIYAMA,2 AND TAKASHI MOURI1
1
Department of Earth and Planetary Sciences, Nagoya University, Chikusa-ku, Nagoya 464-8602, Japan
2
Department of Earth and Environment, Kumamoto University, 2-39-1 Kurokami, Kumamoto 860-8555, Japan

ABSTRACT
A Laser Raman microspectrometry method was applied to metamorphic quartz in quartz-eclogite-,
epidote-amphibolite-, and amphibolite-facies rocks to assess the quantitative correlation between the
Raman frequency shift and metamorphic pressure. Quartz crystals sealed in garnet and other phases
have a higher frequency shift than those in the matrix. Furthermore, the quartz inclusions show a
frequency shift speciÞc to the individual host crystals in eclogites (garnet ≈ kyanite > omphacite ≈
epidote). These observations imply that the residual pressures retained by quartz inclusions depend
on elastic parameters of the host crystals, as discussed by previous researchers. The Raman frequency
shift of quartz inclusions in garnet systematically increases with increasing peak metamorphic pres-
sures from the amphibolite facies (0.30–0.55 GPa/470–570 °C), through the epidote-amphibolite facies
(0.8–1.1 GPa/470–635 °C) to the quartz-eclogite facies (2.1–2.5 GPa/660–710 °C). Calibrations based
on experimental work suggest that the measured Raman frequency shifts signify residual pressures
of 0.1–0.2, 0.4–0.6, and 0.8–1.0 GPa for these three groups of metamorphic rocks, respectively. Nor-
mal stresses (internal pressures) of quartz inclusions in garnet, numerically simulated with an elastic
model, and inferred pressure-temperature conditions at peak metamorphic stage are compatible with
the residual pressures estimated from the frequency shifts. Laser Raman microspectroscopic analysis
of quartz is a simple and effective method for (1) comparison of pressure conditions in metamorphic
rocks formed under various pressure-temperature conditions, and (2) detection of a higher-pressure
signature in metamorphic rocks extensively recrystallized during the subsequent exhumation and
hydration stage.
Keywords: Raman shift, quartz, residual pressure, metamorphism, elastic model

INTRODUCTION rence. Parkinson and Katayama (1999) and Parkinson (2000)

Laser Raman microspectroscopy is a useful tool for iden-
tiÞcation of minute crystals in terrestrial and extraterrestrial
materials and inclusions in optically transmissive host phases
(e.g., Wang et al. 1999; Gillet et al. 2002; Zedgenizov et al.
2004). The Raman spectral investigations of metamorphic rocks
have (1) shown that garnet and zircon are useful pressure con-
tainers for the preservation of ultrahigh-pressure (UHP) phases
(e.g., Sobolev and Shatsky 1990; Sobolev et al. 1995); and (2)
revealed evidence for UHP metamorphism such as coesite and
microdiamond inclusions in zircon and garnet from extensively
retrograded gneissic rocks in several UHP metamorphic terranes
(e.g., Tabata et al. 1998; Parkinson and Katayama 1999; Ye et
al. 2000; Liu et al. 2001).
Coesite is identiÞed by its diagnostic Raman spectrum with
a strong band at about 521 cm–1 at room temperature and atmo-
spheric pressure (Sharma et al. 1981; Boyer et al. 1985; Hemley
1987). The main bands of α-quartz are located at about 464, 205,
and 128 cm–1. These Raman spectra often show shifts to higher
frequency in various degrees depending on their mode of occur-
* E-mail: enami@eps.nagoya-u.ac.jp
reported that (1) the 521 cm–1 band of monocrystalline coesite
inclusions in unfractured zircon and garnet of Kokchetav UHP
metamorphic rocks shifts to 525–526 cm–1; and (2) coesite inclu-
sions occurring inside fractured host crystals show no signiÞcant
Raman frequency shift. They deduced that the monocrystalline
coesite inclusions having the higher Raman frequency shift still
retain ultrahigh overpressures (1.9–2.3 GPa).
Ye et al. (2001) classiÞed the coesite/quartz inclusions in
zircons from the Dabieshan-SuLu UHP metamorphic terrane
into the following three types on the basis of textures: (1) single
grains or aggregates of coesite with polycrystalline quartz totally
enclosed in zircon without fractures; (2) single crystals or ag-
gregates of coesite with quartz that was exposed on the surface
of the section during polishing—these were also enclosed in
zircon; and (3) aggregates of coesite and quartz enclosed in zircon
with fractures. Ye et al. (2001) showed that the Raman spectra
of the coesite and subsidiary quartz inclusions have shifted, and
the amount of shifting was closely correlated with the extent to
which coesite had been converted to quartz. Shifts in the Raman
spectra revealed the following characteristics in the Þrst type of
coesite/quartz inclusions: (1) each coesite/quartz inclusion has a

0003-004X/07/0809–1303$05.00/DOI: 10.2138/am.2007.2438 1303

1304 ENAMI ET AL.: LASER RAMAN MICROSPECTROMETRY OF METAMORPHIC QUARTZ
distinct overpressure, and quartz inclusions tend to retain lower
overpressure than coesite inclusions; and (2) monocrystalline
quartz inclusions give smaller shifts of Raman spectra (lower
overpressures) than polycrystalline quartz inclusions.
Recently, several new applications of the Raman shift for
paleo-geobarometry have been proposed. Izraeli et al. (1999)
reported an internal pressure of 0.13–0.65 GPa in olivine included
in diamond, and calculated source pressures of 4.4–5.2 GPa
for Siberian diamonds considering bulk moduli and thermal
expansion of these two phases. Sobolev et al. (2000) proposed a
“coesite-in diamond” geobarometer that is virtually independent
of temperature, and estimated an initial pressure of 5.5 GPa for
Venezuela diamond formation. Yamamoto et al. (2002) applied
Raman spectroscopic analysis to CO2-dominated ßuid inclusions
in mantle-derived xenoliths and estimated residual pressures as
0.96–1.04 GPa. They additionally suggested that the inclusions
show pressures speciÞc to the individual host minerals of spinel,
orthopyroxene, clinopyroxene, and olivine, similar to the case
of our eclogite samples.
In this contribution, we report systematic measurements of
Raman spectra of quartz crystals in metamorphic rocks recrystal-
lized under quartz-eclogite-, epidote-amphibolite-, and amphibo-
lite-facies conditions. The primary purposes of our study were
to correlate the degree of the frequency shift of quartz crystals
with (1) their modes of occurrence; (2) physical properties of the
host crystals; and (3) metamorphic pressure/temperature (P-T)
conditions of their host rocks. Our results show that the Raman
frequency shift is a powerful and effective tool to (1) estimate
residual pressures retained by minerals; and (2) compare peak
pressure that the metamorphic rocks have experienced.
OUTLINE OF GEOLOGY AND SAMPLE DESCRIPTIONS
Laser Raman microspectroscopic analyses were performed
on three P-T types of metamorphic rocks: (1) quartz-eclogite; (2)
metapelite, metachert, and metabasite of the epidote-amphibolite-
facies from the Sanbagawa metamorphic belt in central Shikoku,
Japan (Fig. 1; Banno and Nakajima 1992; Takasu et al. 1994);
and (3) metapelite of the amphibolite facies from the Altai meta-
morphic belt in northwestern China (Windley et al. 2002; Zheng
et al. 2005). Sample descriptions and representative data set of
mineral compositions are given in the Appendix 1.
Sanbagawa metamorphic rocks
The metamorphic conditions of the Sanbagawa belt can
be discussed in terms of four mineral zones based on mineral
parageneses of metapelites: chlorite, garnet, albite-biotite, and
oligoclase-biotite zones in ascending order of metamorphic
grade, (Fig. 1a; Enami 1983; Higashino 1990). The metamor-
phic grades of the albite-biotite and oligoclase-biotite zones are
equivalent to those of the epidote-amphibolite facies (Enami et
al. 1994). The higher-grade areas in the Besshi region of central
Shikoku are subdivided into the eclogite unit, which is the high-
est structural level of the Sanbagawa belt, and non-eclogite unit
(Fig. 1b; Wallis and Aoya 2000), although the exact position of
the tectonic boundary between the two units is still uncertain.
The eclogite assemblages sporadically occur in the presumed
eclogite unit, and are attributed to crystallization in a deeper
part of the Sanbagawa subduction zone (Takasu 1989; Ota et al.
2004). Lithologies of the eclogite unit and of the surrounding
non-eclogite unit were partially recrystallized under the epidote-
amphibolite-facies conditions during the exhumation and hydra-
tion stage (Aoya 2001; Zaw Win Ko et al. 2005).
Quartz-eclogite facies rocks
The SiO 2 polymorph presumed to have been stable at
peak pressures in the Sanbagawa eclogitic rocks was quartz,
not coesite (e.g., Tsujimori et al. 2000; Aoya 2001; Ota et al.
2004). The eclogites examined for Raman barometry were col-
lected from the eclogite unit in the Besshi region (Fig. 1b), and
grouped into quartz-rich (QE9601 and GO01TT) and quartz-poor
(WK1304 and WK1310) types. Their protoliths are inferred to be
sedimentary mixtures of quartzofeldspathic and basaltic materi-
als for the quartz-rich eclogite, and basalt and/or gabbro for the
quartz-poor eclogite (Banno et al. 1976; Takasu 1989). In these
eclogites, garnet shows a complex zonal structure consisting of a
core overgrown by a discontinuous, narrow mantle (50–200 μm
in width), which are interpreted to have formed at the eclogite
and the later epidote-amphibolite facies stages, respectively.
Quartz is included in the garnet core, but could not be found in
the mantle. The eclogite geothermobarometer (Krogh Ravna and
Terry 2004) gives equilibrium conditions of 2.4 GPa and 680
°C for the quartz-rich eclogite, and 2.1–2.2 GPa and 660–690
°C for the quartz-poor eclogites at the peak metamorphic stage.
Equilibrium P-T conditions of 2.4–2.5 GPa and 665–710 °C were
estimated for kyanite-quartz eclogites collected from the same
area as the quartz-rich eclogites that we studied (A. Miyamoto,
personal communication). Ota et al. (2004) reported T ranging
from 510 to 790 °C and P from 1.4 to 2.5 GPa for the eclogite
assemblages, and attribute the wide range to a P-T gradient in the
eclogite unit. The equilibrium conditions of the present samples
are estimated to correspond to the upper end of the P-T range
reported by Ota et al. (2004).
Epidote-amphibolite facies rocks
Two metapelites (AS1919 and AS2006) and a metachert
(ST22F04b) were collected from the albite-biotite zone of the
Asemi-gawa (River Asemi) area (Higashino 1990; Otsuki and
Banno 1990) and Saruta-gawa area (Kurata and Banno 1974;
Banno 2000) in the inferred non-eclogite unit, respectively
(Fig. 1a). The equilibrium conditions of the albite-biotite zone
were estimated at 0.8–1.0 GPa and 470–590 °C (Enami et al.
1994; Wallis et al. 2000). The garnet-biotite geothermometer
gives an equilibrium temperature of 575 ± 25 °C at 1.0 GPa for
AS2006 (Program GTB: http://ees2.geo.rpi.edu/MetaPetaRen/
Software/GTB_Prog/GTB.html). A metabasite (AS0305) was
sampled from the oligoclase-biotite zone along the Asemi-gawa
area. The sample contains no mineral assemblages suitable for
the conventional calculations of metamorphic P-T conditions.
The equilibrium P-T conditions of the oligoclase-biotite zone
were estimated as 0.9–1.1 GPa/585–635 °C in the Besshi region
(Enami et al. 1994; Wallis et al. 2000), and those of the sample
studied here are probably within this range.
Altai metamorphic rocks
Late Permian metamorphic rocks from the greenschist to
amphibolite facies are widely distributed in the Altai meta-
 ENAMI ET AL.: LASER RAMAN MICROSPECTROMETRY OF METAMORPHIC QUARTZ 1305

FIGURE 1. Metamorphic zone maps of the Sanbagawa metamorphic belt in central Shikoku (a, after Higashino 1990) and the Besshi region (b,
partly modiÞed from Fig. 1 of Aoya 2001) with sample localities (solid circles). Samples with numbers given in regular font were used to compare
Raman frequency shifts with metamorphic pressures (Figs. 6 and 7). Samples with numbers given in italics were used to test the suitability of
Raman barometry for recognizing signatures of higher pressure conditions in metamorphic rocks recrystallized during the subsequent exhumation
and hydration (Fig. 12).

morphic belt. All three Al2SiO5 polymorphs occur in this belt silicates + quartz ± plagioclase are estimated to be 0.30–0.55
(Windley et al. 2002; Zheng et al. 2005), implying conditions GPa and 470–500 °C for sample 148-2, and 0.35–0.55 GPa and
of the medium P/T-type metamorphic facies series. Conditions 520–570 °C for sample 104-9 [Program GTB, THERMOCALC
for the assemblages of garnet + biotite + muscovite + aluminum v3.25 and AX (Holland and Powell 1998)], which are close to
1306 ENAMI ET AL.: LASER RAMAN MICROSPECTROMETRY OF METAMORPHIC QUARTZ

those for the Al2SiO5 triple point of 0.45 ± 0.05 GPa and 550 ±
50 °C (Pattison 1992).

LASER RAMAN MICROSPECTROSCOPY AND RAMAN

SHIFT

Analytical procedures and spectrographic characterization

Raman spectra were obtained with a laser Raman micro-spectrophotometer,
Nicolet Almega XR (Thermo Electron Corp.: gratings 2400 lines/mm), at the
Petrology Laboratory of Nagoya University, equipped with a 532 nm Nd-YAG
laser, charge-coupled-device (CCD) detector (Andro Technology: 256 × 1024
pixels, cooled by a Peltier element) and an automated confocal microscope (Olym-
pus BX51). The objective is an Olympus Mplan-BD 100X (numerical aperture
0.9). Room temperature was kept at 22 ± 1 °C. Spatial resolution is about 1 μm,
and the output power of 25 mW produces an irradiation power of 10 mW on the
samples (R. Harui, personal communication). The pinhole diameter is 25 μm and
the corresponding spectral resolution is about ~2.5 cm–1. Raman spectra were
collected in 6 accumulations of 10 s each. Frequencies of the Raman bands were
calibrated by monitoring the position of a plasma line from the Ne laser (703.241
nm). Using the 520 cm–1 band of the Si wafer, the analytical reproducibility was
checked every day that the analyses were carried out; it was found to be ±0.3 cm–1
(1σ level) over two years.
A (0001) section cut from a euhedral crystal of pegmatitic α-quartz was used
as a standard. At atmosphere pressure and room temperature, the main band for the
α-quartz, which is related to bending vibrations of the intra-tetrahedral O-Si-O angles
(Etchepare et al. 1974), is located at about 464 cm–1 with subsidiary bands of 128
and 205, 355, and 394 cm–1 (Fig. 2). The standard deviation for over 300 duplicate
measurements of the wavenumber of α-quartz is less than ±0.04 cm–1 for the 464
and 128 cm–1 bands and ±0.3 cm–1 for the 205 cm–1 band (Appendix 2 Fig. 1). The
extent of frequency shift of Raman spectra is deÞned by two parameters, ω1 = ν464
– ν205 and ω2 = ν205 – ν128, where νi denotes wavenumber for each band (Fig. 2). We
chose to use differences to avoid systematic errors in measurements of individual νi.
The standard deviation of ω1 and ω2 for the α-quartz standard is ±0.3 cm–1; this level
of precision is largely due to the uncertainty in the ν205 measurement. Temperature
dependencies of ω1 and ω2 are estimated as about –0.06 cm–1/°C (Schmidt and
Ziemann 2000) and –0.03 cm–1/°C (Dean et al. 1982) at room temperature and 0.1
MPa. The experimentally estimated temperature dependencies of ω1 and ω2 are an
order-of-magnitude smaller than the measured standard deviations of ω1 and ω2 (±0.3
cm–1). Thus a room temperature ßuctuation of ±1 °C probably has less inßuence on
the Raman parameters than other factors (Appendix 2 Fig. 1).
The residual pressure retained by quartz inclusions is measured as a function
of the difference in ω1 and ω2 between the α-quartz standard and the sample: ∆ω1
= ωstandard
1 – ωsample
1 and ∆ω2 = ωsample
2 – ωstandard
2 . These two parameters increase with
increasing residual pressure. Most analyzed samples were conventionally pol-
ished thin sections 30–50 μm in thickness. Quartz inclusions in a thicker section
(80–100 μm in thickness) of a quartz-rich eclogite were additionally analyzed
for comparison; this section gave frequency shifts similar to those obtained from
sections of standard thickness.
Mode of occurrences of quartz and Raman shift
Quartz occurs as a matrix phase and/or as an inclusion in
garnet and other constituent minerals. The quartz inclusions are
monocrystalline and mostly 5–20 μm in size. They are usually
spherical or ellipsoidal in form (Figs. 3a and 3b), and in some
cases anhedral or irregular in shape (Figs. 3c and 3d). Follow-
ing Ye et al. (2001), we have identiÞed three types of quartz
inclusions based on their mode of occurrence. The Þrst type of
inclusion is completely surrounded by the host crystal (denoted
as Cs = completely surrounded: Fig. 4a). This type of inclusion
was carefully checked by depth-step scan analysis with the laser
Raman microspectrometer to be sure that it was completely
enclosed in the host crystal and was neither touching the glass
slide nor exposed on the polished surface. The inclusions were
found to be from 5 to 10 μm below the surface in most cases.
The second type of inclusion is either partly exposed on the
FIGURE 2. Representative Raman spectrum of α-quartz at atmosphere
pressure and room temperature, showing deÞnitions of Raman frequency
shifts. Number in parentheses indicates standard deviation (1σ level).
polished surface of host crystal or touches the glass slide (Ex =
exposed: Fig. 4b). No fracture is developed in the two types of
inclusions and surrounding host crystals, and thus these inclu-
sions were deÞnitely encapsulated in their host crystals during
the exhumation stage. The third type of inclusion occurs in a
fractured host crystal in which cracks extend from inside the
host to the inclusion or into and through the inclusion (Fc = in
a fractured host: Figs. 3b and 4c).
Figures 5a and 5b show a depth-step analysis of a Cs inclu-
sion and a horizontal-step analysis of an Ex inclusion in garnet,
respectively. Local increases of the frequency shift of quartz in
the narrow domain extending to 2–4 μm from the quartz-garnet
interface suggest that the mechanical stress concentrates at
this zone, a phenomenon much discussed in the Þeld of mate-
rial science (e.g., Ito et al. 1994; De Wolf 2003). To avoid the
grain-boundary effect on the frequency shift, we will refer only
to analytical data from the central part of crystals.
The quartz inclusions in garnets show systematic variations of
the frequency shifts depending on their mode of occurrence that
are similar to the Raman shift of coesite inclusions in zircon and
garnet reported by Parkinson and Katayama (1999), Parkinson
(2000), and Ye et al. (2000). The shift in frequency for quartz
inclusions in garnet increases in the order Fc, Ex, and Cs (Table
1 and Figs. 6, 7, and 8). The Fc quartz inclusions in garnet and
matrix quartz of these samples show the Raman spectra close to
those of the α-quartz standard. Most fractures in garnet around
the Fc inclusions are likely to have been formed by deforma-
tion during subduction and/or exhumation of the metamorphic
rocks. Thus the Raman spectra of the Fc inclusions in garnet
may express the release of compressive stress (pressure) and
subsequent annealing during the exhumation stage. The Ex
inclusions in garnet probably record the release of compressive
stress in the process of making thin sections. The Cs inclusions
are completely sealed in the host crystals, and thus hold more
certain information about metamorphic conditions than do the
Ex and Fc inclusions in garnet.
In contrast to the case of quartz inclusions in garnet, the
Cs quartz inclusions in staurolite have negative and lower ∆ω1
(–0.9 ± 0.6 cm–1) and ∆ω2 (–0.7 ± 0.7 cm–1) values than the Ex
 ENAMI ET AL.: LASER RAMAN MICROSPECTROMETRY OF METAMORPHIC QUARTZ 1307

FIGURE 3. Photomicrographs
of quartz inclusions in kyanite
(a) and garnet (b, c, and d) from
QE9601 (a and c), GO01TT (b),
and AS2006 (d). Abbreviations are
listed in Table 1.

FIGURE 4. Schematic sections of analyzed samples showing three modes of occurrence of quartz inclusions in a host phase.

FIGURE 5. Step-scan analyses of the Cs quartz inclusion in garnet along vertical (depth) direction (a) and the Ex quartz inclusion in garnet
along-horizontal direction on its surface (b) by the laser Raman microspectroscopy (GO01TT) at room temperature.
1308 ENAMI ET AL.: LASER RAMAN MICROSPECTROMETRY OF METAMORPHIC QUARTZ

TABLE 1. Frequency shifts of Raman bands in spectra of quartz

Rock-types Sample no. Mode Host Frequency shift (in cm–1) GS (μm)
ω1 ω2 NOD
Standard α-quartz // (0001) 259.0 (0.3) 77.8 (0.3) 41

Quartz-eclogite facies ∆ω1 ∆ω2 NA

Qtz-rich eclogite QE9601 Cs inc Grt 9.8 (1.3) 11.2 (1.4) 27 12 (4)
Ex inc Grt 4.1 (0.9) 4.4 (0.9) 13
Fc inc Grt 0.3 (0.4) 0.4 (0.5) 3
Cs inc Ky 9.4 (1.8) 10.7 (2.4) 21
Cs inc Omp 4.0 (2.3) 4.3 (2.4) 6
Cs inc Ep 5.1 (1.8) 5.8 (1.6) 10
Matrix –0.7 (0.7) –0.4 (0.7) 3
Qtz-rich eclogite GO01TT Cs inc Grt 9.1 (2.2) 10.4 (2.7) 30 16 (7)
Ex inc Grt 5.6 (1.4) 6.2 (1.4) 18
Fc inc Grt 1.7 (0.3) 2.2 (0.1) 2
Matrix –0.1 (0.6) 0.1 (0.5) 19
Qtz-poor eclogite WK1304 Cs inc Grt 8.9 (1.2) 10.0 (1.3) 19 15 (5)
Ex inc Grt 3.7 (1.2) 4.2 (1.1) 6
Fc inc Grt –0.3 (0.6) 0.0 (0.6) 3
Matrix –0.3 (0.4) –0.1 (0.4) 3
Qtz-poor eclogite WK1310 Cs inc Grt 9.7 (1.1) 11.1 (1.1) 14 n.m.
Ex inc Grt 5.9 (1.3) 6.4 (1.6) 7
Fc inc Grt 0.0 (0.2) –0.2 (0.2) 2
Matrix Grt –0.2 (0.3) –0.4 (0.4) 3

Epidote-amphibolite facies
Metapelite AS1919 Cs inc Grt 5.8 (1.8) 6.3 (1.9) 5 6 (2)
Ex inc Grt 2.4 3.4 1
Matrix –0.2 (0.7) –0.2 (0.7) 3
Metapelite AS2006 Cs inc Grt 7.1 (0.8) 8.1 (1.0) 12 10 (4)
Ex inc Grt 3.8 (1.6) 4.6 (1.3) 8
Fc inc Grt 0.2 (1.5) 0.2 (1.0) 2
Matrix 0.7 (0.4) 0.8 (0.4) 3
Metachert ST22F04b Cs inc Grt 7.0 (0.7) 8.0 (0.9) 42 8 (2)
Ex inc Grt 3.6 (1.2) 4.3 (1.3) 28
Fc inc Grt 0.0 (0.7) 0.5 (0.6) 11
Matrix –0.5 (0.3) –0.3 (0.2) 5
Metabasite AS0305 Cs inc Grt 4.5 (1.0) 5.1 (1.0) 13 n.m.
Ex inc Grt 2.5 (1.0) 2.7 (1.1) 7
Fc inc Grt –0.2 –0.2 1
Matrix 0.3 (0.7) 0.2 (0.8) 2

Amphibolite facies
Metapelite 104-9 Cs inc Grt 0.8 (0.4) 1.0 (0.5) 6 19 (4)
Ex inc Grt –0.2 (0.5) 0.0 (0.6) 6
Matrix –0.4 (0.4) –0.3 (0.3) 7
Metapelite 148-2 Cs inc Grt 1.6 (0.6) 1.7 (0.7) 12 20 (9)
Ex inc Grt 0.0 (0.3) –0.1 (0.2) 6
Cs inc St –0.9 (0.6) –0.7 (0.7) 10
Ex inc St –0.2 (0.8) –0.1 (0.8) 10
Matrix 0.0 (0.3) –0.2 (0.1) 4
Notes: Grt = garnet; Ky = kyanite; Omp = omphacite; Ep = epidote; St = staurolite; NOD = Number of operating days; NA = Number of analyses; GS = grain size of
inclusion; n.m. = not measured. Abbreviations for mode of quartz and parameters (ω1, etc.) are described in the text. Number in parentheses indicates standard
deviation (1σ level).

inclusions (∆ω1 = –0.2 ± 0.8 cm–1 and ∆ω2 = –0.1 ± 0.8 cm–1) and
matrix quartz (∆ω1 = 0.0 ± 0.3 cm–1 and ∆ω2 = –0.2 ± 0.1 cm–1)
(Table 1 and Fig. 9). The inverse relationships could indicate that
the Cs inclusions in staurolite retain tensile stress (e.g., Ito et al.
1994; De Wolf 2003), and the relatively higher frequency shift
of the Ex inclusion in staurolite implies release of the tensile
stress in the process of making thin section.
Host crystals and Raman shift
The Cs quartz inclusions show systematic variations of the
frequency shift depending on their host crystals (Fig. 10). In a
quartz-rich eclogite (QE9601), the Cs quartz inclusions in garnet
and kyanite have values similar to each other: ∆ω1 = 9.8 ± 1.3
cm–1 and ∆ω2 = 11.2 ± 1.4 cm–1 in garnet, and ∆ω1 = 9.4 ± 1.8
cm–1 and ∆ω2 = 10.7 ± 2.4 cm–1 in kyanite. The Cs inclusions in
omphacite and epidote, on the other hand, have lower frequency
shifts than those in garnet and kyanite: ∆ω1 = 4.0 ± 2.3 cm–1 and
∆ω2 = 4.3 ± 2.4 cm–1 in omphacite and ∆ω1 = 5.1 ± 1.8 cm–1 and
∆ω2 = 5.8 ± 1.6 cm–1 in epidote. The bulk modulus (κ298) of garnet
(166–175 GPa: Wang and Ji 2001) and kyanite (156–193 GPa:
Comodi et al. 1997; Yang et al. 1997; Winkler et al. 2001a) are
similar to each other, and are higher than those of omphacite
(114–131 GPa: Bhagat et al. 1992; Pavese et al. 2001; Nishihara
et al. 2003) and epidote (106–136 GPa: Ryzhova et al. 1966;
Winkler et al. 2001b). In a metapelite of the amphibolite facies,
the Cs quartz inclusions in staurolite with a low bulk modulus
(124.6 GPa: Aleksandrov and Prodaivoda 1993) have lower
frequency shifts than that in garnet as described in the previous
section (Fig. 9). The positive correlations between the frequency
shifts of Cs quartz inclusions and elastic parameters of their host
crystals imply that the variations of Raman shift of quartz can be
explained based on an elastic model as discussed below.
ENAMI ET AL.: LASER RAMAN MICROSPECTROMETRY OF METAMORPHIC QUARTZ 1309

FIGURE 6. Correlation between the Raman

frequency shifts (∆ω1 and ∆ω2) of quartz and their
modes of occurrence at room temperature for the
quartz-eclogite-facies rocks from the eclogite unit of
the Sanbagawa metamorphic belt. Squares indicate
data from thicker section and other symbols show
data from normal sections.

FIGURE 7. Correlation between the Raman

frequency shifts (∆ω1 and ∆ω2) of quartz and
their modes of occurrence at room temperature
for the epidote-amphibolite-facies rocks from the
non-eclogite unit of the Sanbagawa metamorphic
belt.
1310 ENAMI ET AL.: LASER RAMAN MICROSPECTROMETRY OF METAMORPHIC QUARTZ

FIGURE 8. Correlation between the

Raman frequency shifts (∆ω1 and ∆ω2) of
quartz and their modes of occurrence at
room temperature for the amphibolite-facies
rocks from the Altai metamorphic belt.

Raman barometry
The degree of Raman shift for quartz is intimately connected
to its mode of occurrence (Figs. 6, 7, and 8) and physical proper-
ties of the host crystals (Figs. 9 and 10). Sobolev et al. (2000)
reported that smaller coesite inclusions in diamond give a higher
Raman shift than the larger ones. Wendt et al. (1993) numeri-
cally discussed a quantitative dependence between grain-size
and conÞning pressure of α-quartz included in almandine-rich
garnet. Accordingly, we discuss the relationships between the
Raman frequency shifts of quartz inclusions and metamorphic
P-T conditions using the data of Cs inclusions in garnet with
regard to its grain size (Fig. 11). The frequency shifts of the
Cs inclusions from the quartz-eclogite facies rocks show a
weak negative correlation with inclusion size, and are 3.8–12.7
and 4.8–14.8 cm–1 for ∆ω1 and ∆ω2, respectively. The epidote-
amphibolite- and amphibolite-facies samples do not show an
obvious correlation between the frequency shifts and grain sizes,
and have frequency shifts of ∆ω1 = 5.2–8.5 and ∆ω2 = 6.1–9.5
cm–1, and ∆ω1 = 0.7–2.3 and ∆ω2 = 0.4–2.4 cm–1, respectively,
systematically lower than the quartz-eclogite facies rocks.
FIGURE 9. Comparison of the Raman frequency shifts (∆ω1 and ∆ω2)
Raman spectra of quartz show a systematic pressure-dependent
of the Cs and Ex quartz inclusions in staurolite, Ex quartz inclusion in
shift in series of experiments by Dean et al. (1982), Hemley (1987), garnet, and matrix quartz of an amphibolite facies metapelite (148-2) at
Liu and Mernagh (1992), and Schmidt and Ziemann (2000). The sys- room temperature. Abbreviations are listed in Table 1.
tematic differences of frequency shifts shown in Figure 11 certainly
reßect those of residual pressures retained by quartz. Applying the
calibrations of Liu and Mernagh (1992) and Schmidt and Ziemann The simplest model considers a spherical inclusion in an
(2000), the maximum frequency shifts (∆ω1 and ∆ω2) give residual inÞnite isotropic medium with an external stress applied at in-
pressures of 0.8–1.0, 0.4–0.6, and 0.1–0.2 GPa for the quartz- Þnity on the medium (Van der Molen 1981; Van der Molen and
eclogite-, epidote-amphibolite- and amphibolite-facies samples, Van Roermund 1986). Elastic parameters of minerals are taken
respectively (cf. Fig. 11 and Appendix 2 Fig. 2). The estimated as constants and their dependence on temperature and pressure
residual pressures systematically increase in the ascending order of is neglected. Dependence of stresses on the grain size of the
P and P-T conditions at the peak metamorphic stages. inclusion does not appear in this model.
The Þnite size model takes into account the grain size effects
Numerical simulation of internal pressures of both the inclusion and its host mineral on stresses in and around
Normal stresses (residual pressures) in and around a spheri- the inclusion (Gillet et al. 1984). The dependence of elastic pa-
cal inclusion have been formulated theoretically as a “spherical rameters on temperature and pressure is also considered.
inclusion model” (e.g., Van der Molen 1981). Several varieties The three-shelled composite sphere model considers all
of the formulation have been applied to the problem of coesite spherical radii as parameters (Lee and Tromp 1995; Nishiyama
retention in garnet from UHP metamorphic rocks (Gillet et al. 1998). The model can be extended to n-spherical shells (arbitrary
1984; Van der Molen and Van Roermund 1986; Nishiyama 1998). number of shells). Elastic parameters are taken as constants. A
They have several features as summarized below. volume change due to phase transition (or metamictization) is
 ENAMI ET AL.: LASER RAMAN MICROSPECTROMETRY OF METAMORPHIC QUARTZ 1311

FIGURE 10. Comparison of the Raman frequency shifts (∆ω1 and ∆ω2)
of the Cs quartz included in garnet, kyanite, omphacite, and epidote of
a quartz-rich eclogite (QE9601) at room temperature.

FIGURE 11. Correlations between the Raman frequency shifts

considered, although thermal expansion is neglected.
From among these models we will adopt the simplest one
because host garnets include several of inclusions that makes
the exact application of the Þnite-size model difÞcult or even
meaningless, and because all the inclusions in this study can be
considered as originally included not as coesite but as quartz.
According to Van der Molen (1981) the internal pressure PQtz
(normal stress) of a spherical inclusion of quartz in garnet, which
has undergone temperature change ∆T under the external pressure
PGrt on the garnet, can be written as
(∆ω1 and ∆ω2) of the Cs quartz inclusion in garnet and its grain size
at room temperature. Grain size (d) was calculated as d = a × b ,
where a and b are, respectively, the major and minor axes of quartz
inclusions approximated by an ellipsoid. Data for irregularly shaped
quartz inclusions such as shown in Figures 3c and 3d are not shown in
this Þgure. Abbreviations are Qtz-ecl = quartz-eclogite-facies sample;
Ep-amp = epidote-amphibolite-facies sample; Amp = amphibolite-facies
sample; LM92 = Liu and Mernagh (1992); SZ00 = Schmidt and Ziemann
(2000). Pressure scale was calculated based on the pressure dependence
of the Raman frequency shifts shown in Appendix 2 Figure 2.

κQtz
PQtz =
κGrt (3κQtz + 4μ Grt )
{ P (3κ
Grt Grt
+ 4μ Grt ) − 4κGrt μ Grt ΔT ΔA}
TABLE 2. Bulk (κ) and shear (μ) moduli and thermal expansion pa-
rameter (A) at standard condition (10–4 GPa and 298 K)
where κ denotes bulk modulus, μ is shear modulus, and ∆A stands Mineral κ (GPa) μ (GPa) A (K–1)
for the difference of thermal expansion parameters between Quartz 37.8* 2.38E-05‡
quartz and garnet deÞned as ∆A = AGrt – AQtz. Based on this equa- Almandine 175.1† 92.1† 1.57E-05‡
Pyrope 170.1† 90.2† 1.98E-05‡
tion, we calculated the internal pressures of quartz inclusions for Grossular 166.3† 98.1† 1.63E-05‡
four end-members of a garnet host: almandine, pyrope, grossular, Spessartine 171.8† 93.3† 1.71E-05‡
and spessartine. Table 2 summarizes parameters used in our * Bass (1995).
† Wang and Ji (2001).
calculations. Elastic parameters are taken from Bass (1995) and ‡ Fei (1995).
Wang and Ji (2001), and thermal expansion parameters from Fei
(1995). Our concept of the calculations is such that the system from 0.68 to 0.85 GPa in the case of almandine, 0.78 to 0.95
has experienced a temperature change from the peak temperature GPa for pyrope, 0.38 to 0.49 GPa for grossular, and 0.71 to
to 298 K under the external pressure, and that the inclusion will 0.88 GPa for spessartine. Because the host garnet in the studied
retain the internal pressure at atmospheric pressure after removal samples is almandine-rich with moderate amounts of pyrope and
of the external pressure. Therefore, we used the magnitude of grossular components, the calculated results for almandine and
thermal expansion parameters at room temperature that will pyrope hosts are consistent with the residual pressure of 0.8–1.0
describe the misÞt between the inclusion and the host causing the GPa measured by Raman spectroscopy (Fig. 11). As for the
internal pressure at the Þnal state (at room temperature and the epidote-amphibolite-facies rocks in the non-eclogite unit (Ppeak
external pressure PGrt). Thermal expansion parameters at room = 0.8–1.1 GPa and Tpeak= 470–635 °C), the calculations give the
temperature were calculated according to an equation given in internal pressure of 0.19–0.31 GPa for almandine, 0.26–0.38
Fei (1995, Table 1). GPa for pyrope, 0.08–0.15 GPa for grossular and 0.21–0.33 GPa
The calculated results of normal stress at standard conditions for spessartine. The garnet in the metachert is spessartine-rich,
are listed in Table 3. The Sanbagawa eclogites (Ppeak = 2.1–2.5 with small or moderate amounts of almandine and grossular
GPa and Tpeak = 660–710 °C) show the internal pressure ranging components, and that from metapelite and metabasite is a solid
1312 ENAMI ET AL.: LASER RAMAN MICROSPECTROMETRY OF METAMORPHIC QUARTZ

TABLE 3. Numerical simulations of internal pressures of metamorphic quartz included in garnet at standard condition
Sanbagawa Qtz-rich eclogite Qtz-poor eclogite
Tpeak (°C) 665 710 660 690
Ppeak (GPa) 2.4 2.5 2.4 2.5 2.1 2.2 2.1 2.2
PQtz (GPa)
in Alm 0.81 0.85 0.80 0.84 0.69 0.73 0.68 0.72
in Prp 0.90 0.95 0.90 0.94 0.78 0.82 0.78 0.82
in Grs 0.46 0.49 0.45 0.48 0.39 0.41 0.38 0.41
in Sps 0.84 0.88 0.83 0.87 0.72 0.76 0.71 0.75

Sanbagawa epidote-amphibolite facies rocks

Olig-biotite zone Ab-biotite zone
Tpeak (°C) 585 635 470 590
Ppeak (GPa) 0.9 1.1 0.9 1.1 0.8 1.0 0.8 1.0
PQtz (GPa)
in Alm 0.23 0.31 0.22 0.30 0.22 0.30 0.19 0.27
in Prp 0.30 0.38 0.30 0.38 0.27 0.36 0.26 0.34
in Grs 0.10 0.15 0.09 0.14 0.10 0.15 0.08 0.13
in Sps 0.25 0.33 0.24 0.32 0.23 0.31 0.21 0.29

Altai amphibolite facies rocks

Tpeak (°C) 470 570
Ppeak (GPa) 0.30 0.55 0.30 0.55
PQtz (GPa)
in Alm 0.02 0.12 –0.01 0.09
in Prp 0.07 0.17 0.06 0.16
in Grs –0.02 0.04 –0.04 0.02
in Sps 0.04 0.13 0.01 0.11
Notes: Tpeak and Ppeak = deduced peak metamorphic conditions; PQtz = estimated internal pressure at standard condition; Alm = almandine; Prp = pyrope; Grs =
grossular; Sps = spessartine; Olig = oligoclase; Ab = albite.

solution of almandine, spessartine, and grossular. Incorporation

of the grossular component will decrease the internal pressure
drastically, but the effect may not be very large in the case of the
Sanbagawa samples because the grossular contents are less than
30 mol%. The calculated results for the epidote-amphibolite-fa-
cies samples give an internal pressure 0.1–0.3 GPa less than the
internal pressure measured by Raman spectroscopy (0.4–0.6
GPa, Fig. 11). Finally in the case of the Altai metapelites (Ppeak
= 0.30–0.55 GPa and Tpeak = 470–570 °C), our calculation gives
–0.01 to 0.12 GPa for almandine, 0.06–0.17 GPa for pyrope,
–0.04 to 0.04 GPa for grossular, and 0.01 to 0.13 GPa for spes-
sartine. Because the Altai garnets are almandine-rich with small
amounts of grossular component (less than 8 mol%), the calcu-
lated results are consistent with the measured residual stresses of
0.1–0.2 GPa (Fig. 11). In summary, elastic modeling can predict
the residual pressure around quartz inclusions in garnet, at least
as a Þrst approximation.

Use of residual pressure to identify high-metamorphic

pressures
Because the residual pressures retained by quartz inclusions
increase with increasing peak metamorphic pressures, they can
be used to identify rocks that had undergone metamorphism
under high pressures when other indicators are absent or have
been destroyed. For example, systematic measurements of Ra-
man spectra of quartz crystals implies that Cs quartz inclusions
in garnet with ∆ω1 > 8.5 ± 0.3 and ∆ω2 > 9.5 ± 03 cm–1 at room
temperature are inferred to be relics of quartz-eclogite-facies
FIGURE 12. The Raman frequency shifts (∆ω1 and ∆ω2) of the Cs
metamorphism in the Sanbagawa belt (Figs. 6, 7, and 11). Most
quartz inclusions in garnet of metapelite (ZWK07a) and metabasites
of rocks in the Sanbagawa eclogite unit (Fig. 1b) were completely (IR2001, IR21N03g, and WK1305) from the inferred eclogite unit of
recrystallized during the exhumation and hydration stages, and the Sanbagawa belt (cf. Fig. 1b) at room temperature. These samples
thus have no phase diagnostic of eclogite-facies metamorphism were completely recrystallized under the epidote-amphibolite-facies
such as omphacite. To Þnd evidence for high-pressure metamor- conditions, and preserve no phase diagnostic of the eclogite-facies stage.
phism in such rocks, we applied Raman barometry to a metapelite Abbreviations are listed in Figure 11.
 ENAMI ET AL.: LASER RAMAN MICROSPECTROMETRY OF METAMORPHIC QUARTZ
(ZWK07a) and metabasites (IR2001, IR21N03g, and WK1305)
in which there is no other evidence of eclogite-facies metamor-
phism. The metapelite (ZWK07a) contains quartz, garnet, para-
gonite, phengite, barroisite, chlorite, albite, and rutile as major
matrix phases, and its equilibrium conditions at early high-pres-
sure stage were estimated as 1.8 GPa and 520–550 °C based on
the assemblage of chloritoid, barroisite, and paragonite included
in garnet (Zaw Win Ko et al. 2005). The metabasites (IR2001,
IR21N03g, and WK1305) consists mainly of barroisitic-taramitic
amphibole, garnet, epidote, phengite, paragonite, plagioclase,
quartz, and rutile. The Cs inclusions in garnet record high ∆ω1 and
∆ω2 values, i.e., 10.8 and 11.9 cm–1 in ZWK07a, 10.6 and 11.7
cm–1 in IR2001, 10.3 and 11.4 cm–1 in IR21N03g, and 10.3 and
11.6 cm–1 in WK1305, respectively (Fig. 12). Thus, it is almost
certain that the four samples have experienced quartz-eclogite-
facies metamorphism, demonstrating that Raman spectrometry is
a powerful and useful tool for recognizing signatures of higher-
pressure conditions in metamorphic rocks recrystallized during
the subsequent exhumation and hydration.
ACKNOWLEDGMENTS
We are deeply indebted to S. Dunn (Mount Holyoke College), H. Kagi (Uni-
versity of Tokyo), I.-M. Chou (U.S. Geological Survey), and E.S. Grew (University
of Maine) for their careful reading and constructive suggestions, which led to
signiÞcant improvements in the manuscript, and appreciate Y. Hirahara (Nagoya
University), S. Wallis (Nagoya University), and two anonymous reviewers for
commenting on the earlier versions of this manuscript. We also thank C.Q. Zheng
(Jilin University) for giving the loan of the Altai samples, and to R. Harui (Thermo
Fisher ScientiÞc K.K.) for giving approval to quote the unpublished data. This
study was supported Þnancially in part by Grant-Aid for ScientiÞc Research from
JSPS (M.E., 14540448 and 18340172).
REFERENCES CITED
Aleksandrov, K.S. and Prodaivoda, G.T. (1993) Elastic properties of minerals.
Crystallography Reports, 38, 214–234.
Aoya, M. (2001) P-T-D path of eclogite from the Sambagawa belt deduced from
combination of petrological and microstructural analyses. Journal of Petrol-
ogy, 42, 1225–1248.
Banno, S. and Nakajima, T. (1992) Metamorphic belts of Japanese islands. Annual
Review of Earth and Planetary Sciences, 20, 159–179.
Banno, S., Yokoyama, K., Iwata, O., and Terashima, S. (1976) Genesis of epidote
amphibolite masses in the Sanbagawa metamorphic belt of central Shikoku.
Journal of Geological Society of Japan, 82, 199–210 (in Japanese with
English abstract).
Banno, Y. (2000) Intermediate high-pressure exhumation of the northern segment
of the Sanbagawa belt, Saruta-gawa area, central Shikoku, Japan. Lithos,
50, 289–303.
Bass, J.D. (1995) Elasticity of minerals, glasses, and metals. In T.J. Ahrens, Ed.,
Mineral physics and crystallography: A handbook of physical constants, 2, p.
45–63. American Geophysical Union, Washington, D.C.
Bhagat, S.S., Bass, J.D., and Smyth, J.R. (1992) Single-crystal elastic properties of
omphacite-C2/c by Brillouin spectroscopy. Journal of Geophysical Research,
B, Solid Earth and Planets, 97, 6843–6848.
Boyer, H., Smith, D.C., Chopin, C., and Lasnier, B. (1985) Raman microprobe
(RMP) determinations of natural and synthetic coesite. Physics and Chemistry
of Minerals, 12, 45–48.
Comodi, P., Zanazzi, P.F., Poli, S., and Schmidt, M.W. (1997) High-pressure
behavior of kyanite: Compressibility and structural deformations. American
Mineralogist, 82, 452–459.
De Wolf, I. (2003) Raman spectroscopy: about chips and stress. Spectroscopy
Europe, 15, 6–13.
Dean, K.J., Sherman, W.F., and Wilkinson, G.R. (1982) Temperature and pressure
dependence of the Raman active modes of vibration of α-quartz. Spectrochi-
mica Acta A, 38, 1105–1108.
Enami, M. (1983) Petrology of pelitic schists in the oligoclase-biotite zone of the
Sanbagawa metamorphic terrain, Japan: phase equilibria in the highest grade
zone of a high-pressure intermediate type of metamorphic belt. Journal of
Metamorphic Geology, 1, 141–161.
Enami, M., Wallis, S.R., and Banno, Y. (1994) Paragenesis of sodic pyroxene-
bearing quartz schists: implications for the P-T history of the Sanbagawa belt.
Contributions to Mineralogy and Petrology, 116, 182–198.
1313
Etchepare, J., Merian, M., and Smetankine, L. (1974) Vibrational normal modes of
SiO2. I. α and β quartz. Journal of Chemical Physics, 60, 1873–1876.
Fei, Y. (1995) Thermal expansion. In T.J. Ahrens, Ed., Mineral Physics and
Crystallography: A Handbook of Physical Constants, 2, p. 29–44. American
Geophysical Union, Washington, D.C.
Fukura, S., Mizukami, T., Odake, S., and Kagi, H. (2006) Factors Determining the
Stability, Resolution, and Precision of a Conventional Raman Spectrometer.
Applied Spectroscopy, 60, 946–950.
Gillet, P., Ingrin, J., and Chopin, C. (1984) Coesite in subducted continental crust:
P-T history deduced from an elastic model. Earth and Planetary Science Let-
ters, 70, 426–436.
Gillet, P., Sautter, V., Harris, J., Reynard, B., Harte, B., and Kunz, M. (2002) Ra-
man spectroscopic study of garnet inclusions in diamonds from the mantle
transition zone. America Mineralogist, 87, 312–317.
Hemley, R.J. (1987) Pressure dependence of Raman spectra of SiO2 polymorphs;
α-quartz, coesite, and stishovite. In H. Manghnani Murli and Y. Syono, Eds.,
High-pressure research in mineral physics, 39, p. 347–359. American Geo-
physical Union, Washington, D.C.
Higashino, T. (1990) The higher-grade metamorphic zonation of the Sambagawa
metamorphic belt in central Shikoku, Japan. Journal of Metamorphic Geol-
ogy, 8, 413–423.
Holland, T.J.B. and Powell, R. (1998) An internally consistent thermodynamic data set for
phases of petrological interest. Journal of Metamorphic Geology, 16, 309–343.
Ito, T., Azuma, H., and Noda, S. (1994) Stress measurements in silicon substrates
with TiSi2 patterns using Raman microprobe. Japanese Journal of Applied
Physics, 33, 171–177.
Izraeli, E.S., Harris, J.W., and Navon, O. (1999) Raman barometry of diamond
formation. Earth and Planetary Science Letters, 173, 351–360.
Kretz, R. (1983) Symbols for rock-forming minerals. American Mineralogist, 68,
277–279.
Krogh Ravna, E.J. and Terry, M.P. (2004) Geothermobarometry of UHP and HP eclog-
ites and schists; an evaluation of equilibria among garnet-clinopyroxene-kyanite-
phengite-coesite/quartz. Journal of Metamorphic Geology, 22, 579–592.
Kurata, H. and Banno, S. (1974) Low-grade progressive metamorphism of pelitic
schists of the Sazare area, Sanbagawa metamorphic terrain in central Shikoku,
Japan. Journal of Petrology, 15, 361–382.
Lee, J.K.W. and Tromp, J. (1995) Self-induced fracture generation in zircon. Journal
of Geophysical Research B: Solid Earth and Planets, 100, 17753–17770.
Liu, J., Ye, K., Maruyama, S., Cong, B., and Fan, H. (2001) Mineral inclusions in
zircon from gneisses in the ultrahigh-pressure zone of the Dabie Mountains,
China. Journal of Geology, 109, 523–535.
Liu, L.G. and Mernagh, T.P. (1992) High-pressure Raman study of the α-quartz
forms of SiO2 and GeO2 at room temperature. High Temperatures-High Pres-
sures, 24, 13–21.
Nishihara, Y., Takahashi, E., Matsukage, K., and Kikegawa, T. (2003) Thermal
equation of state of omphacite. American Mineralogist, 88, 80–86.
Nishiyama, T. (1998) Kinetic modeling of the coesite-quartz transition in an elastic
Þeld and its implication for the exhumation of ultrahigh-pressure metamorphic
rocks. The Island Arc, 7, 70–81.
Ota, T., Terabayashi, M., and Katayama, I. (2004) Thermobaric structure and
metamorphic evolution of the Iratsu eclogite body in the Sanbagawa belt,
central Shikoku, Japan. Lithos, 73, 95–126.
Otsuki, M. and Banno, S. (1990) Prograde and retrograde metamorphism of he-
matite-bearing basic schists in the Sanbagawa belt in central Shikoku. Journal
of Metamorphic Geology, 8, 425–439.
Parkinson, C.D. (2000) Coesite inclusions and prograde compositional zonation
of garnet in whiteschist of the HP-UHPM Kokchetav massif, Kazakhstan: A
record of progressive UHP metamorphism. Lithos, 52, 215–233.
Parkinson, C.D. and Katayama, I. (1999) Present-day ultrahigh-pressure condi-
tions of coesite inclusions in zircon and garnet: Evidence from laser Raman
microspectroscopy. Geology, 27, 979–982.
Pattison, D.R.M. (1992) Stability of andalusite and sillimanite and the Al2SiO5
triple point; constraints from the Ballachulish aureole, Scotland. Journal of
Geology, 100, 423–446.
Pavese, A., Diella, V., Levy, D., and Hanßand, M. (2001) Synchrotron X-ray pow-
der diffraction study of natural P2/n-omphacites at high-pressure conditions.
Physics and Chemistry of Minerals, 28, 9–16.
Ryzhova, T.V., Aleksandrov, K.S., and Korobkova, V.M. (1966) The elastic proper-
ties of rock-forming minerals; V, Additional data on silicates. Physics of the
Solid Earth, 2, 63–65.
Schmidt, C. and Ziemann, M.A. (2000) In-situ Raman spectroscopy of quartz; a
pressure sensor for hydrothermal diamond-anvil cell experiments at elevated
temperatures. American Mineralogist, 85, 1725–1734.
Sharma, S.K., Mammone, J.F., and Nicol, M.F. (1981) Raman investigation of ring
conÞgurations in vitreous silica. Nature, 292, 140–141.
Sobolev, N.V. and Shatsky, V.S. (1990) Diamond inclusions in garnets from metamor-
phic rocks; a new environment for diamond formation. Nature, 343, 742–746.
Sobolev, N.V., Shatsky, V.S., Vavilov, M.A., and Goryaynov, S.V. (1995) Zircon
in high-pressure metamorphic rocks in folded regions as a unique container of
inclusions of diamond, coesite, and coexisting minerals. Transactions of the
1314 ENAMI ET AL.: LASER RAMAN MICROSPECTROMETRY OF METAMORPHIC QUARTZ
Russian Academy of Sciences. Earth Science Sections, 336, 79–85.
Sobolev, N.V., Fursenko, B.A., Goryainov, S.V., Shu, J., Hemley, R.J., Mao, H.K.,
and Boyd, F.R. (2000) Fossilized high pressure from the Earth’s deep interior:
the coesite-in-diamond barometer. Proceedings of National Academy of Sci-
ences, USA, 97, 11875–11879.
Tabata, H., Maruyama, S., and Shi, Z. (1998) Metamorphic zoning and thermal
structure of the Dabie ultrahigh-pressure-high-pressure terrane, central China.
The Island Arc, 7, 142–158.
Takasu, A. (1989) P-T histories of peridotite and amphibolite tectonic blocks in
the Sambagawa metamorphic belt, Japan. In J.S. Daly, R.A. Cliff, and B.W.D.
Yardley, Eds., The Evolution of Metamorphic Belts, 43, p. 533–538. Blackwell
ScientiÞc Publications, Oxford.
Takasu, A., Wallis, S.R., Banno, S., and Dallmeyer, R.D. (1994) Evolution of the
Sambagawa metamorphic belt, Japan. Lithos, 33, 119–133.
Tsujimori, T., Tanaka, C., Sakurai, T., Matsumoto, M., Miyagi, Y., Mizukami,
T., Kugimiya, Y., and Aoya, M. (2000) Illustrated introduction to eclogite in
Japan. Bulletin of Research Institute of Natural Sciences, Okayama University
of Science, 26, 19–40.
Van der Molen, I. (1981) The shift of the α-β transition temperature of quartz
associated with the thermal expansion of granite at high pressure. Tectono-
physics, 73, 323–342.
Van der Molen, I. and Van Roermund, H.L.M. (1986) The pressure path of solid
inclusions in minerals: the retention of coesite inclusions during uplift. Lithos,
19, 317–324.
Wallis, S. and Aoya, M. (2000) A re-evaluation of eclogite facies metamorphism
in SW Japan: Proposal for an eclogite nappe. Journal of Metamorphic Geol-
ogy, 18, 653–664.
Wallis, S., Takasu, A., Enami, M., and Tsujimori, T. (2000) Eclogite and related
metamorphism in the Sanbagawa belt, Southwest Japan. Bulletin of Research
Institute of Natural Sciences, Okayama University of Science, 26, 3–17.
Wang, A., Jolliff, B.L., and Haskin, L.A. (1999) Raman spectroscopic characteriza-
tion of a Martian SNC meteorite; Zagami. Journal of Geophysical Research,
E, Planets, 104, 8509–8519.
Wang, Z. and Ji, S. (2001) Elasticity of six polycrystalline silicate garnets at pres-
sure up to 3.0 GPa. American Mineralogist, 86, 1209–1218.
Wendt, A.S., D’Arco, P., Goffé, B., and Oberhänsli, R. (1993) Radial cracks around
α-quartz inclusions in almandine; constraints on the metamorphic history of the
Oman mountains. Earth and Planetary Science Letters, 114, 449–461.
Windley, B.F., Kröner, A., Guo, J., Qu, G., Li, Y., and Zhang, C. (2002) Neopro-
terozoic to Paleozoic geology of the Altai orogen, NW China: New Zircon age
data and tectonic evolution. Journal of Geology, 110, 719–737.
Winkler, B., Hytha, M., Warren, M.C., Milman, V., Gale, J.D., and Schreuer, J.
(2001a) Calculation of the elastic constants of the Al2SiO5 polymorphs andalu-
site, sillimanite and kyanite. Zeitschrift für Kristallographie, 216, 67–70.
Winkler, B., Milman, V., and Nobes, R.H. (2001b) A theoretical investigation of
the relative stabilities of Fe-free clinozoisite and orthozoisite. Physics and
Chemistry of Minerals, 28, 471–474.
Yamamoto, J., Kagi, H., Kaneoka, I., Lai, Y., Prikhod’ko, V.S., and Arai, S. (2002)
Fossil pressures of ßuid inclusions in mantle xenoliths exhibiting rheology of
mantle minerals; implications for the geobarometry of mantle minerals using mi-
cro-Raman spectroscopy. Earth and Planetary Science Letters, 198, 511–519.
Yang, H., Downs, R.T., Finger, L.W., Hazen, R.M., and Prewitt, C.T. (1997)
Compressibility and crystal structure of kyanite, Al2SiO5, at high pressure.
American Mineralogist, 82, 467–474.
Ye, K., Yao, Y., Katayama, I., Cong, B., Wang, Q., and Maruyama, S. (2000) Large
areal extent of ultrahigh pressure-metamorphism in the Sulu UHP terrain of
Eastern China: new implications from coesite and omphacite inclusions in
zircon of granitic gneiss. Lithos, 52, 157–164.
Ye, K., Liou, J.G., Cong, B., and Maruyama, S. (2001) Overpressures induced by
coesite-quartz transition in zircon. American Mineralogist, 86, 1151–1155.
Zaw Win Ko, Enami, M., and Aoya, M. (2005) Chloritoid and barroisite-bearing
pelitic schists from the eclogite unit in the Besshi district, Sanbagawa meta-
morphic belt. Lithos, 81, 79–100.
Zedgenizov, D.A., Kagi, H., Shatsky, V.S., and Sobolev, N.V. (2004) Carbonatitic
melts in cuboid diamonds from Udachnaya kimberlite pipe, Yakutia; evidence
from vibrational spectroscopy. Mineralogical Magazine, 68, 61–73.
Zheng, C.Q., Xu, X.C., Enami, M., and Kato, T. (2005) Monazite ages and
geological implications of andalusite-sillimanite type metamorphic belt in
Aletai, Xinjiang, China. Global Geology, 24, 236–242 (in Chinese with
English abstract).
MANUSCRIPT RECEIVED AUGUST 30, 2006
MANUSCRIPT ACCEPTED MARCH 9, 2007
MANUSCRIPT HANDLED BY EDWARD GREW
APPENDIX 1: PETROLOGY OF SAMPLES STUDIED
Representative mineral compositions are listed in Appendix
Table 1.
Sanbagawa quartz eclogite
Sample QE9601 is composed mainly of garnet (Alm45–60
Prp21–39Sps1–3Grs11–25), omphacite (Jd44–48Aug45–54Aeg0–10), phen-
gite (Si = 3.27–3.35 per formula unit: pfu for O = 11), kyanite,
epidote [YFe = Fe3+/(Al + Cr + Fe3+) = 0.17–0.20], and quartz,
with small amounts of rutile, apatite, and zircon. [Abbreviations
of end-members follow Kretz (1983) other than Aeg (aegirine).]
Rutile needles (1–2 × 10–30 μm) have exsolved from the core
part of zoned garnet. Glaucophane (Al = 1.50–1.61 pfu and Na
= 1.81–1.82 pfu for O = 23) is included in garnet. Omphacite
is rimmed by secondary symplectitic aggregates of barroisite
(Si = 6.57–7.61 pfu and Ca = 0.63–1.48 pfu) + albite (An5–9),
whereas kyanite is replaced by Þne-grained mixtures of phengite
and paragonite. Secondary titanite rims matrix rutile. It is not
clear whether paragonite was stable at the peak eclogite stage:
it occurs both as inclusions in garnet and as isolated grains in
the matrix.
Sample GO01TT has constituent minerals and mineral
chemistries similar to those of QE9601, except that kyanite is
absent.
Sample WK1304 consists mainly of garnet (Alm 43–55
Prp8–16Sps1–12Grs29–37), omphacite (Jd34–44Aug50–66Aeg0–6), phengite
(Si = 3.29–3.32 pfu), epidote (YFe = 0.12–0.16), and quartz with
accessory rutile, titanite and apatite. Pargasite-hornblende/bar-
roisite (Si = 6.26–6.68 pfu, Ca = 1.28–1.65 pfu) is included in
garnet. Slightly Si-richer amphiboles occur as independent grains
(Si = 6.73–7.46 pfu and Ca = 1.34–1.77 pfu) and symplectitic
aggregates (Si = 6.88–7.49 pfu and Ca = 1.60–1.76 pfu) with
albite (An1–3) in the matrix.
Sample WK1310 is composed mainly of garnet (Alm52–60
Prp9–17Sps2–7Grs23–29), omphacite (Jd39–47Aug47–56Aeg0–6), pheng-
ite (Si = 3.29–3.32 pfu), epidote (YFe = 0.14–0.18), and quartz
with accessory rutile and apatite. Barroisitic amphibole (Si =
6.68–7.15 pfu and Ca = 1.05–1.41 pfu) occurs as inclusions in
garnet and omphacite. Isolated amphibole grains in the matrix
have pargasitic-katophoritic compositions (Si = 6.03–6.76 pfu
and Ca = 1.19–1.64 pfu) and symplectitic amphibole aggregates
with albite (An1–9) have edenitic-actinolitic compositions (Si
= 6.79–7.30 pfu and Ca = 1.66–1.73 pfu). Secondary titanite
rims rutile.
Sanbagawa epidote-amphibolite-facies rocks
Metapelite (AS2006) consists of garnet (Alm39–70Prp1–9
Sps1–41Grs19–33), biotite [XMg = Mg/(Mg + Fe) = 0.48–0.50, TiO2
= 1.3–1.4 wt%], chlorite (XMg = 0.47–0.55), phengite (Si =
3.24–2.38 pfu), epidote, pargasite/hornblende (Si = 6.41–6.98
pfu and Ca = 1.51–1.72 pfu), albite (An1–3), titanite, rutile, apatite
and fully ordered graphite.
Sample AS1919 has constituent minerals and mineral chem-
istries similar to those of AS2006, except that amphibole and
biotite are absent and tourmaline is present.
Metachert (ST22F04b) is composed predominantly of
quartz with subordinate amounts of garnet (Alm9–23Prp3–9Sps55–75
Grs13–15)], winchite (Si = 7.18–7.37 pfu and Ca = 0.65–0.93 pfu),
epidote (YFe = 0.28–0.29)/piemontite, phengite (Si = 3.27–3.37
pfu) and albite and accessory ilmenohematite and apatite.
Metabasite (AS0305) is mainly composed of taramite (Si
= 6.21–6.49 pfu and Ca = 1.32–1.49 pfu), garnet (Alm57–71
 ENAMI ET AL.: LASER RAMAN MICROSPECTROMETRY OF METAMORPHIC QUARTZ 1315

APPENDIX TABLE 1. Representative analyses of major constituent minerals

Quartz-eclogite facies Epidote-amphibolite facies Amphibolite facies
Qtz-rich eclogite Qtz-poor eclogite Metapelite Metabasite Metapelite
QE9601 WK1310 AS2006 AS0305 148-2
Grt Cpx Ms Grt Cpx Ms Grt Bt Amp Chl Grt Amp Ms Grt Bt Ms Pl
SiO2 39.8 55.0 49.1 38.8 55.4 49.1 37.4 37.2 42.7 26.1 37.8 42.2 46.5 37.5 36.8 46.1 56.5
TiO2 0.01 0.21 0.45 0.19 0.14 0.36 0.04 1.41 0.41 0.06 0.04 0.47 0.66 0.00 1.53 0.49 0.00
Al2O3 22.5 11.1 28.4 21.9 10.6 29.8 21.1 15.9 15.0 20.9 20.7 15.3 31.3 21.3 19.0 34.7 27.4
Cr2O3 0.02 0.00 0.03 0.01 0.00 0.00 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.
FeO* 22.9 6.47 2.69 25.4 6.54 2.60 30.0 19.9 16.3 23.9 31.3 17.2 3.81 27.2 14.1 1.56 0.28
MnO 0.41 0.03 0.00 1.48 0.20 0.02 0.48 0.09 0.05 0.00 1.14 0.07 0.00 7.80 0.15 0.00 0.00
MgO 9.28 7.30 3.02 4.43 7.39 2.72 2.20 11.1 8.24 16.1 3.53 8.86 1.64 3.80 13.5 0.68 0.02
BaO n.a. n.a. 0.11 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. 0.68 n.a. n.a. n.a. n.a.
CaO 6.02 11.9 0.02 8.77 12.3 0.01 7.95 0.00 9.87 0.01 5.67 9.42 0.01 2.67 0.00 0.02 9.51
Na2O n.a. 7.26 1.01 n.a. 7.07 1.06 n.a. 0.07 3.19 0.00 n.a. 3.03 1.61 n.a. 0.35 1.30 6.30
K2O n.a. 0.00 9.98 n.a. 0.00 9.81 n.a. 9.55 0.48 0.00 n.a. 0.45 8.78 n.a. 9.07 9.54 0.04
Total 100.94 99.27 94.81 100.98 99.64 95.48 99.17 95.22 96.24 87.11 100.18 97.00 94.99 100.27 94.50 94.39 100.07
* Total iron as FeO. Abbreviations are Qtz = quartz; Grt = garnet; Cpx = clinopyroxene; Ms = white mica; Bt = biotite; Amp = amphibole; Chl = chlorite; Pl = plagio-
clase; n.a. = not analyzed.

Prp4–14Sps1–12Grs14–29), epidote (YFe = 0.27–0.28), muscovite (Si

= 3.13–3.26 pfu), oligoclase, quartz, ilmenite, dolomite/calcite,
rutile and apatite. Titanite occurs as inclusions in garnet. Second-
ary chlorite (XMg = 0.51) partly replaces garnet.

Altai amphibolite facies rocks

Sample 104–9 consists mainly of garnet (Alm64–71Prp5–12
Sps10–21Grs6–8), biotite (XMg = 0.51–0.54 and TiO2 = 1.3–1.6 wt%),
muscovite (Si = 3.06–3.13 pfu), staurolite (XMg = 0.14–0.18 and
ZnO = 0.23–0.41 wt%), andalusite, chlorite (XMg = 0.53–0.55)
and quartz with accessory ilmenite and apatite.
Sample 148-2 is composed mainly of garnet (Alm60–63
Prp12–15Sps17–20Grs5–8), biotite (XMg = 0.62–0.63 and TiO2 = 1.3–1.6
wt%), muscovite (Si = 3.07–3.12 pfu), staurolite (XMg = 0.25–0.26
and ZnO = 0.10–0.22 wt%), kyanite, chlorite, plagioclase
(An36–55) and quartz with accessory ilmenite and apatite.

APPENDIX 2 FIGURE 1. Changes in the Raman frequency shifts of

(a) 464, (b) 205, and (c) 128 cm–1 bands and (d) room temperature with
APPENDIX 2: BASIC DATA OF RAMAN SHIFT OF
time. S = Start-up of the Raman spectral instrument; R = Completion of
α-QUARTZ CCD detector cooling and Raman measurement start.
The stability of Raman spectra was veriÞed by continuous
measurement of α-quartz standard over Þve hours (Appendix
2 Fig. 1). The frequency shifts of the 464 and 128 cm–1 bands
gradually decreased by a small amount during the Þrst hour of
measurement, and then were fairly constant over the remain-
ing four hours. The 205 cm–1 band has a broader peak width
and consequently a larger standard deviation (±0.3 cm–1 for 1σ
level) of the frequency shift than the 464 and 128 cm–1 bands
(±0.03–0.04 cm–1). Fukura et al. (2006) carefully considered
factors determining the stability, resolution, and precision of a
conventional Raman spectrometer in their laboratory, and re-
ported that a change of 0.4 °C in room temperature causes peak
positions of ruby ßuorescence spectrum to shift by 0.052 cm–1.
In our case, variation in room temperature of 0.7 °C seems to
have but a minor effect on the frequency shifts considering the
analytical precision of 0.03–0.3 cm–1.
APPENDIX 2 FIGURE 2. Frequency shifts (∆ω1 and ∆ω2) in the Raman
spectrum of α-quartz as a function of pressure at room temperature. Data
are from Schmidt and Ziemann (2000) = solid lines and Þlled diamonds;
Liu and Mernagh (1992) = dashed line.